The article is devoted to the actual problem of increasing the complexity of user passwords in systems with remote access to increase their information security. It proposes a method based on the integrated application of the input characters of the password and the time intervals between them. In the article, using the inhibitor time network Petri, a model of a dynamic password setting process was constructed, on its basis the process of forming a reference image and user authentication is described. The above calculations of increasing the complexity of the password prove the effectiveness of the proposed approach.

Keywords: authentication, password, dynamic process, Petri net, complexity, password retrieval

The article is devoted to the description of the method of forming steganographic network messages using the ICMP service protocol for their hidden transfer, bypassing restrictions of firewalls and other computer network protection systems. It describes the functions of the ICMP protocol, its advantages when used in steganography tasks, and the specifics of processing ICMP packets by operating systems and intermediate network equipment. Special attention is paid to the practical implementation and application of the proposed method. It describes the network utilities for working with packages and an example of their use for transmitting a hidden message. For the software implementation of the network steganography method, it is proposed to use the C # programming language and the SharpPCap and Packet.NET libraries, for which the article contains examples of use and the necessary parameters for forming packets with steganographic messages.

Keywords: information security, network steganography, ICMP protocol, TCP / IP stack, encapsulation, SharpPCap

The paper is devoted to the actual problem of ensuring the information security of web-sites. It discusses a method for detecting intentional threats to the confidentiality of information as a result of unauthorized access and is manifested in the form of atypical requests to resources by users. The paper proposes a method based on data mining. Its essence lies in the classification of users' behavior on the basis of information about their actions using an artificial neural network. As a basis for the implementation of the proposed tools, site security audit tools are used as a source of information. The structure of neural network, training methods and practical application are described, and the effectiveness of the proposed methods is evaluated.

Keywords: Data mining, artificial neural network, web-design, machine learning, classification

The article is devoted to the problem of using artificial neural networks as one of the methods of data mining (intellectual data analysis) for classifying the competencies of university students when determining the educational specialization. It describes the complexity of the process of selecting specialization and the associated negative consequences, as well as an approach to solving the given problem using software classification tools. As a basis for the implementation of the proposed tools, the student's information system is used on the website of the University of Tartus, Syria. The article presents data selection criteria to form the training sample, which includes the academic performance in some courses as input vectors. The values of the output vector depend on completed specialization and it get into the training set only if it is properly selected. On the basis of these data, the structure of a multilayer artificial neural network is formed and the learning algorithm is selected, the results of which are reflected on the university's website in the form of advice on the choice of future specialization, which has allowed increase the effectiveness of the educational process.

Keywords: Data mining, artificial neural network, web-design, machine learning, classification

The article is devoted to the problem of data protection from interception as a result of the "man in the middle" attacks. The proposed technique for detecting these attacks is based on the analysis of the headers of transit packets passing through the default gateway. Based on the data obtained, a table of correspondence between IP and MAC addresses is constructed, for which software provides up-to-date and reliable information. The addresses of packets passing through the gateway are compared with the records in this table and, in case of a mismatch and impossibility of confirming the correctness of addresses in the headers of the channel and network layers, it is concluded that there is an additional intermediate node in the network that appeared as a result of the default gateway substitution. The article presents approaches to software implementation of this technique, describes the packet analysis algorithm.

Keywords: local area network, man-in-the-middle, DHCP-spoofing, ARP-poisoning, traffic analysis, gateway, network address, packet, ARP-table

It describes a method of train set forming for using artificial neural networks to search inauthentic rows  in databases tables. An existing methods of the reliability ensure involve the use of integrity constraints and provides a truthfulness, but there is still a possibility of entering of inauthentic data, appropriate to all constraints. A more accurate assessment of reliability is possible with the use of artificial neural networks that require a training set. The main requirement for the training set - representation includes sufficiency, diversity and evenness. The approaches to each of these requirements are describes. Also calculations a sufficient number of rows for training neural networks of various types is given, as well as the results of experiments that confirm correctness of the theoretical calculations.

Keywords: database, authenticity, artificial neural networks, training set, representation

Offers intelligent method of assessing the reliability of data (verification), in databases, based on the theory of artificial neural networks. The method consists of three steps: data clustering using the Kohonen network, training radial basic network and its computation using the validation database table rows. To create a radial basis networks and their training Backpropagation developed software tool. The proposed verification method of data can be used in various applications of information systems.
Keywords: database, relational algebra, verification, radial basis neural network training neural networks, Kohonen networks, clustering.

Keywords: